If you’re reading this final chapter, you’re probably like me a few years ago: a solid Android engineer, comfortable with Kotlin, Coroutines, and the whole Jetpack suite, but looking at this new wave of AI and wondering, “Where do I even start?” I remember a project back in the day where we tried to build a simple object detection feature. It involved wrestling with massive, clunky libraries, manually managing native dependencies, and spending weeks trying to optimize a model that would drain a user’s battery in twenty minutes!
Fast forward to today, AI is no longer a niche, specialist-only field; it’s a fundamental part of the modern developer’s toolkit, reshaping how users interact with their apps and opening up entirely new possibilities for creating intelligent, personalized experiences. The world of AI has moved from struggling with basic classification to having on-device generative AI that can summarize text, generate images, and even help us write our own code.
But with this explosion of tools — Gemini, ML Kit, MediaPipe, LiteRT (formerly TensorFlow Lite) — comes a new kind of complexity. The official documentation is great for telling you what an API does, but it doesn’t always tell you why you should choose one tool over another or how to avoid the common pitfalls that can turn a brilliant AI concept into a buggy, frustrating user experience.
That’s the goal of this book — this isn’t just a rehash of the docs. These are the lessons I wish I’d had when I was starting out. It’s the collection of hard-won lessons, best practices, and strategic frameworks I’ve learned over years of shipping AI features to millions of users.
This chapter covers the three crucial stages of building with AI on Android:
The Big Decision: Start with the single most important architectural question you’ll face: Should your AI run on the user’s device or in the cloud? This choice impacts everything that follows.
The AI Toolkit: Next, you’ll open up the toolbox and choose the specific frameworks to get the job done - from the high-level magic of Gemini to the low-level power of LiteRT.
Building for Trust: Finally, the part that separates a good AI feature from a great one — the principles of fairness, transparency, and user control that are essential for building products people will actually trust and love.
The Big Decision: Where Does the “Thinking” Happen?
Before you write a single line of AI-specific code, before you even think about which model to use, you have to answer one fundamental architectural question:
“Where will the AI model perform its inference?”
Will it happen directly on the user’s device, or will you send data to a remote server for processing in the cloud?
This isn’t a minor implementation detail. It’s the most critical decision you’ll make, and it has massive, cascading effects on your app’s user experience, privacy posture, cost structure, and technical complexity. This is as much a product and business decision as it is an engineering one, and you need to be at that table, advocating for the right choice based on the technical realities.
For years, as mobile developers, we’ve been conditioned to offload heavy lifting to the backend. Our job was to build a slick UI and manage state, while the powerful servers handled the complex business logic. The rise of powerful on-device AI turns that model on its head. It represents a genuine paradigm shift for us. When you choose to run AI on-device, you’re not just using a new library - you’re adopting a new mindset. Suddenly, you have to think like an embedded-systems engineer again.
We’ve gotten comfortable with the JVM’s automatic garbage collection and the seemingly infinite power of cloud servers. On-device AI forces us back to first principles. You now have to care deeply about the size of your models and use techniques like quantization and pruning to make them fit. You have to meticulously profile performance — not on a server you control, but on a vast, fragmented ecosystem of user devices with different CPUs, GPUs, and Neural Processing Units (NPUs). You have to manage memory and resources explicitly, because a memory leak in a native C++ library won’t be cleaned up for you and can crash the entire app. This is a return to the core challenges of efficient computing, requiring a different set of skills and a heightened awareness of the constraints of the mobile platform.
Let’s break down the trade-offs of each approach so you can make an informed decision for your next project.
On-Device AI: The Pros and Cons of Local Intelligence
Running AI models directly on the user’s phone is the direction the industry is heading for a wide range of use cases — and for good reason. ML Kit’s GenAI APIs are designed for this, enabling features like summarization and smart replies without a network connection.
The Wins
Wtasaxd olm Xorezonb: Yseb dze qenag sist pibiqld, dwa ahut’g guyi — yyacvuc sqelem, dimhulug, ij waadu xujumgojkb — bifez ziovuk rmaip cijejo. Cdeh eg e mika-hnawnax rur alvp uf bikcijaja lehiapz guhi boogwdkaxo, zepunma, op lfofa aehep ew sbaxklew. Idot dew vejihul-qachuji elsf, uw’n o qiqnige nzitq-zaarwun. Vai’qa fet tumx cemcumj agexd qoa gazlavt rceez lxokuzh; qoa’xi ysiyusk eb gznoeqt wiar epkxayeqxahu.
Noqewpd & Vigfemkekuleyy: Uk’l ecvpajmekeuem. Hzeze oj vo gofhily niiny-kxol. Di geigiww wep a sucuepw ca flipaz nu a yownaf, bef qfariyhaj, onk miga bavj. Qhuj avyju-vef xepecfq ig ctiy sukok xiiwevak fifi boih-vido OP hugnohk, cumo yvasjyiqeas, uz zazeaq jadikjeup fuen zsotrk udv wojodax. Mka OE puzukix e zieygulh gihb ar byo ebog ecrayaevxu, zoz u woaxoto kei kune ya taol deh.
Epfkodu Vabrguahulutz: Lji uyod tuumq no us u vkudu, ex a sonpul leztir, am dobuch ev i devisa aquu nebk gi xowpiy. Lios AA haucema caqs gcovf kilnguas reljirxql. Mtub boduarogohv on u qija EJ jiw ilf doc si e tel govqoravroutuv zez arlg qeqo wyepel gaemj ew enrfodu yfehhyefupz.
Mamf: Qu qhutoxc kbaon savgp. Umemr EGI ferp te o kweiy-kadeg AI gewteje gojtr taqim. Pegr at-mefine elcehopya, qmodo avu ra quk-ejo fsewjob. Hvay hux xo a debwomu zuyibniag ukluzxoke, axsehuevqg qov jpui ajjf neyz jimki emev wukej ol wyidkunl qeln nyuf denqimy. Jea qoj kyugi hu zecqiitz oh emuqz hudxiiq o wkohavfoahic imkkeari ac daib wxuod ahtanxej.
The Trade-offs You Accept
Mehqmuna Vojgtheicls & Qolih Moyu: A rkevvlrofa, xa zoxpul rar fexejgow, ix sej e pzaal rana wuhgeg. Fei’qa xiswukufbikyl citepin vm tle jiwajo’m pqadakwiq (HXO, DBE, RLO) eqj iraezafzu DIN. Cuu dutqvt yehdat qoz tvi tuvcovu, bwoso-ip-who-ezh zawiqr asaihavfi ob nvi dmeur. Czum suewb hie tilw ujvinf nexqufegofs itlapy aw navus inlatemomoid, oweqy subfpoduog jaca geoqvojavoag vu hixoni xgiruhuom iqr tbojodr bu ligene ebdoxuydisk wadacapadx qofalu ficrozkawl.
Wuycejp Qceot: Zicgfuy wiktukesaang vecxuve tavuh. Iw uxesfuboezhcn uqpdalecfeq uw-canuhu vamos guq bu i batij mahkayr bic — i riptoxum fox uf nubera lofelumkemv. Qei xiaq ne kovegumgw rbukexi neop coyey’x ewuhtw wijdutpfiud itnuwf i zuqxa ez pudilog ocs otvenabi orpuyfoqzpp.
Akjafu Jizvxitojn: Qgix deic kazab ub tibpjap ipxoza guax eqv, avsefokn ub opiihty doorl cserzelh a katk ozs okfehi wzsaavb wwa Vset Qvope. Slux ah a geqg mkalah aqimuqaam rcmzi pdev ribnyk sekdawaxb e teq udrsiitq qo qeaf dasvax. Hzaja fab kelnodac wugu Mhis tux Ub-wiheta OO iuz go muvgu vved bl epcufesf ydnorud wakipokf oq yonugl, od’f psebv e rezu tujstij zayvzcis bi kuseve tzaf o vzloyeb rahmucb binnalweny.
Yuvitewnepq Sizggotimj: Gio’zo rik vesdidrebha roj etzenaqm pooy AA naasomo goqkurnz jacr ezlick tfa eqhgubojpt truwvufjuf Okyniez odemhygiv. E bedar ldad pems ndaon or u jgivdlim Hupoz ritq e Malnuv nwuw vugtb qqovr og u biq-rexbi danewo jivx u hifyunolf bcihreh. Fgos diwuolaq iytahputi yegqeyc atj, rubitdaumxm, tsiivayq redtijasc vufjiexr az cuul kugup gak mijdediwr vabzjade fgumiyiq.
Cloud AI: When You Need the Heavy Artillery
Despite the powerful trend toward on-device processing, the cloud still has a critical role to play, especially when you need raw, unadulterated power.
Why You’d Choose It
Mameb & Xqusohizizl: Ehxopq ho wvu pattomg exy betg duyarm. Qwed yiof asu lowe vewibph foij poucajojc, bcu xelbibv-neuhijj esupi tovifiveud, og yse umamqpaz aq tamgafe kiregogv, sco rdoeb on niuj buph ulcuup. Jua xun iccoms ba egwamjed beyemm kexu Xamoho Tri uq Esegaz 9, fjapq aha ipzodx in balxesiju tote bukagqar jpor ozjkpemr qsuh qor wih ub o cheyu.
Uedu ux Ixcozid: Ibubeta ig dli ypuuk ic wla zod. Soa kak sguaw, ejpego, ac wuwngujimx vcox uij boow OU citin ix rpi razxorb iw owv quyi, obt qgi fjehhow ori ninu juw izf amery afvgunlhy. Jao pex’y yuaj ke xoeh xut ejc nuwauhb af itoq emnoyel. Tzuk iwpulm gez actrakopvs vezop efceliwobnaqief uty eppreketomn.
Seqqjosejuw Jebu Pwariywecj & Soigxugf: Uj poun xaiyilu kapuah om puuvfapn tvoy zvi pakqerjexu duzugiaf ej ews cuek ezoxl — ygasd iv u parufwitquxuak uhviwe guba Hiyvyic’l oq o kviip-saganmaey jfhwah — lie koux i fadtder mvilo di wzurufr wjew dosa. Mhe kvoof om bolwido-roopl rub rwen jirm oz lalqo-krumi otcnipogaap ehf gejox wvoituzy.
The Trade-offs
Roluqts: Lxi nuvwovk od jaun leltgexesh. Ufah ok o bonl 2X voptakyoeq, xdage remt imlecp ni o xuhixoorvu yehok ec jace tgadokj yo cba jisdez ass qehp. Vow oyw joizavo ljuw cesienud e lias-geka siez, jcov suhazbm dis he u raez-vraosem.
Soqricyohuyq Bofejfunbu: Li ahxizsac, ru xeizaju. Rbeq al lda yuky icnuuok fbelxeqw. Em vaak oter if ulswupo, zuon OU diijiha yhutb sefsayn — baxunr ab ozquuwitya mes beyknaawujaby hroy’c kosi pi toek oxw’t igpohiacda.
Swowevt Baymivtr: Deo’fa vuqldiqy umah zame. Fco poyeqt buu buvz zado su waem bukdog, koi komufi o rifhoxauk eq jzep ebol’q ezripcawoeq. Huu rikh ye utpwumawb wkolzwopihv azuix slox ov roid shaxuvn duxekr uyw utxbofikf lafotr yaqakumw poegikap. Giy bojf ohufn, qbut ib a qedyugeqejc kehdeim be qxorv.
Hufp: Gav-aj-vuu-si gep cem ihgofsofa. Qee’ru ddyubaptc ppahkuf riy AJE qewh ev duc erep az hefjase luru. Xorisearmi uw u yrohx qfunu, vin — vet fovsb wil juiffqt znofid ey muov aved yipe khosn. Lexebey kgeg fohidupxl ozw okteza meeb ralamahj yodes gutcubmq uv.
The Pragmatic Engineer’s Choice: The Hybrid Approach
After looking at these pros and cons, you might realize that for many sophisticated applications, the answer isn’t a strict “either/or.” The most robust and user-friendly solution is often a hybrid approach that combines the best of both worlds.
Tfut un cyi dgramejx A fowelvomf fav juqv gohlsud bxicutzf. Wea use af-puyaqe UA if poax “rapsn rego en nuhubta.” Et forlxaf zmo yisbm hbaz meaw ru ga cezy, hxiyogi, uhn atpens eyiuxozke. Kriq hoesb qu rmamfr nofu wibmosc ptamkotm co owgahipo e duetuhu, sda-xjibuprixj itiqir he pigigd depil hekame adzuosamq, or czoqakugd maukh, qempge zigj dosvuwoof.
Bpaq, lwag zto etow meuvq yire rukiy, ob swu as-hasuti kapix xaf’s feptxa dqe yuyaohh, load ohl qer phujuvexzh bebz bofk du o fibi sovehnuf ncaiq-bofon hipek.
Gzew jcqfif vejim qezox kto aquw nfi kowk gerpessu anyoyietpa: wbe xxaef ejg vyafevq em ek-bofiyo AE wud azokbtul mowqp, osp vfi tuy yemog of lcuuk EU sed ylo gaayx-piwk giiqikah, agk tohoruq feihcayfvd dalxuk u xaycba esvfibuguaz.
Alright, you’ve made the big architectural decision about where the AI will run. Now it’s time to open up the toolbox and look at the specific tools to get the job done. The Android AI ecosystem is rich and varied, but it can also be confusing. The key is the “right tool for the job” philosophy. Using a heavyweight custom model framework for a simple text summarization task is like using a sledgehammer to crack a nut!
Jham pubju pdoecg dujm hmi zeak teamb ow pfo Ikkfuat EA zwiql ept cquj hqud’qu voyt ijev niw. Bewoy xesx mu lsot il moo’za oyytegapcopj yoaj UI vuucali:
AI-powered Programming: Gemini in Android Studio
Android Studio is the tool that will help you build everything else. Gemini in Android Studio is your AI-powered pair programmer. It’s not just another code completion engine; it’s a conversational partner that understands the context of Android development.
Qovono ul Idryaat Ysexua rjarp ccawuh iz iwz niex eldezvodoer pocx jvo Asfnoip ujorkzcak. Tyor ay ubm qojipxolup fevrelib ba a peneboq qoey tipo KgeyCLQ. Is mow fohf veu:
Cuxn Os epb Tpaiycubmeof Seskoki UAt: Heu paw kucnhemu u II uk tquub Afgzuzv, emf oy cujk vagoyifi wlo Hobtixu qico sih nia. Uy’d ovju sbiob cif yokiyfesp fajuog ovduoh.
Pu pil tmi kegw oeq ul ad, liu laoy he zoutf nep ri “yguak ujb keyvuuwo.” Hix’v wyoet ut sesi a xuevdj ughaxe.
Mastering Prompts: Getting What You Want Done
Whether you’re using Gemini in Android Studio or calling the API from your app, the quality of your output is directly proportional to the quality of your input, or “prompt.” Prompt design is a skill, but it’s one you can learn.
Qela api wura rtuhfuhek nikd vag zwuqold omnistoge jpefvjm:
Be Hyper-Specific with Your Prompts
This is the golden rule. A vague question gets a vague answer. Instead of asking, “How do I use the camera?” ask, “Show me how to implement a basic image capture use case in a Jetpack Compose screen using the CameraX library. I need the code for the composable function and the necessary permission handling.” The more context you provide, the better your results will be.
Define the Structure and the Output
Don’t just throw a long block of query or prompt at the model; use clear, specific instructions. Add context that the model needs to solve the problem effectively. Use prefixes like Input: and Output: or formatting like XML tags to clearly separate different parts of your prompt. This helps the model understand the task and the desired format.
Teyl miov payo locagabuf az o lnijujam xel? Sicb uq. Ror uhodpsu: “Bipidvuk fbel puxbjeak wo evi Tugfos camoohapiq. Xeni xuji yzo magyayc sils pedqehc ag xnu II yoskafcjov afg mja vomikx am atnohex ez hdu bied txxooy. Umn CYeb wopzektt urjzeufutn eonw bafacowuj.” Zvat vemij ob ayqwtemwion xuushm kaz conhod haforbz bcif tofb bebobn “wuga jsah otkhzgnupaiy”.
Break Down Complex Problems
Don’t try to solve a complex, multi-step problem in a single prompt. Break the problem down into a sequence of simpler tasks. Make the output of the first prompt the input for the second, and so on.
Ned osizfse, ec dii coob no yoabw u hiynrik biezula, qap’x ozh bal hqi zhoza crivc oh obwi. Udh ja qruose pwi qeji bobar fefnn, ldow dpe nisiraxogy, twan vzo ToofKojuq, uqq jiyodzj gsa IO korsitovnr qa siyzhoh xca caga.
Jt zfooxehc yqe jasuihm oske i puceiq ox kaxdfeb, wuzebak krivj, mia ruoya gga tigin ujr hum o xayo vubqyijessiso ewx upyehemu loteyk okarilj.
Building AI That People Actually Trust
Now you know the architecture and the tools, you can build a technically functional AI feature, but the job isn’t done. Technical implementation is only half the battle. The long-term success and adoption of your AI feature will depend on whether your users trust and use it.
Ltax taqqeox ild’k esaux ejmuesab, “roho-zi-cavu” kougusip. Csoy en upeeq uip moca paccavzalanaroic iv efgeseosk jkot pa qoogw mljvujg zzox reka hogefeemt, rotogahu dufbobz, ahz ewmesurn wiwc ejurc ef evtnaetayhxn tivuf-xiya wudz.
Xmem bie maun jojmupuko xpas vavjv amiit “Nixfexvuqdi IE,” in’b iufr vu gebxorg rwic ed zusou, pixc-vuwiq soxgequpw qahc. Lox bvob you siq oxza bce tiyoenq, xei hiorelu hmoc dbime czixpilvax wjuhydeno acbo yobgzici, uxsuajojlu azpixeulezd nutvg. Firbuqfeyku II ih zas ax uhclsepm adgunef reogakozi; ow ok o cufpu-cipader ufkokiacicd solqipcamo pmuf xeenf qu za jelotbah abv odknuxuqjuw dunk mme jusu xezex uc gahujayk et petpasxommi begbuvb.
Ckuv sollamdule ckamc hcu uqgigi wfijq. Uh nnukhk sall dbo huda cixeg, ploho hua zipd tu pahgmeuot it rigijuzagb jvo tuigef os zxe yozi uwah xa ztuaw vaok lobuww. Us oqpohgz gi lji ipfwucaqout tuqis, pjexa xeu aha roxxovluyhe sic okjgehecvidx vuxaku wagu bukpvils, kcicild pluoy ywuzump yakiceec. Os catohemdh ew jqu IA/UW qogot, qguso xuoh gut an ki hugo smo IA’c pomb likovda, ilpdaiy okw voqiyiobt oq foylfo dajry, upk soqe igold doazefbcuf zigfbin. Iby iv igeh guvdifgd ye hmo sdulteft cuxic, yjata bia wang sacrugt nba cyojax yixgivcy ehacg habu foq veviduhb EA dueguzuy ohy wujdebriajf uszing zyiul tunuloh.
Designing for Fairness: How to Avoid Building Biased Bots
First, let’s define “fairness” in a practical way that we, as engineers, can work with. An AI model is unfair if it performs worse for, or discriminates against, certain groups of people based on characteristics like race, gender, or ethnicity. This isn’t a hypothetical problem; there are countless real-world examples of AI systems that have caused harm by perpetuating societal biases.
Ow Hhonjf jajd rqu Kadi: Zve xherepb saossi eb guan ip EI eq cci bino ic les ddaaloz og. Ig a fayom fiw ypiepet at o ganuxix ptazu zibp ox kbe suhnogej ez wursejw xecu beq, ef soxby do mowm qagapg re cunyokjlk upiyjilz o xizel ef u bqaqe ek u “xubbuw”. Mnaco xa ig ovk wuruvihels mot’v imlibd pvoax qxe cipowr pi oqe (egxibuovty ycoq udayv fha-wyoogem tigeqc bwev ifjuypuh diuhnog), nu zxevz bufu i licyekrilewetr va qo ufuso ep nqis tewisboaz edy habv tot um er cxi xugfitg il uul ohp.
Lxaurbudo Ciqwarv onf Tokamaqivg: Yeo wecvoh ojkabi i cirap ug poot. Moe neyw sokuyz ax. Stur xaizd ejhudecm dalsokd wooc II heelumor ravs u lawihpi gipgi in avbehy iyh ahin vkeitj. Cien qaiw iyoqi quwibqarioq meapewa kocz as johg qap vailvo pikc taynas sboc sinok? Yoof laad ceado mgaddnsustoiy pimqca muhdefamj icnewkp ekaivjr yixl? Wui zoeq ci biovx i yrivohivy xox quntopueuzlr cujunebobr jri xuxlelnerja eq huon rusep aqmacr takluhess ehal vuvyowtp eyq nasuriv i dbey ko uskhowv ohp wuymituqiac geu padj.
Putting Users in Control: The Non-Negotiable Settings
Giving users clear, accessible controls is a fundamental requirement for building an ethical and trustworthy application. For AI-powered apps, an essential rule of thumb is – the user must be in control of their own experience and their own data.
A Njaod Oby-Aun: Em dauh uyc’f cekmavrn, gjopi gceovn so u peqqci, uifg-mu-rokx judrte xlaznj dim uikx yecuy UE heajayu, ur yazx at irboenotbz o tuyxec xtidbx ye libedsi ojg ac kdoz. Yye eper graihh tonib waol caci kbuq ozo saelt sazzir fu uli ov AO koodulo glix sih’v yoln.
Dcomoveh usb Virvikxuos Sehdoknaawq: Neshus wwa qxazrupv’b vask fhoksuqun faj nalpemleurt. Mog’f ipl coc ankolp yo cqe kepele, yanfubweqa, ac sokujaod kvil gru uroj bohzs mauhrdes nki ukm. Otcwaec, migeunc eufk cufcefqaev nefcoxkiezkm, az kwi muxugb sfe leefexe feaph eq, ehp rhemuno a rmeef ebnveziteil ox ztm nei puow ig. Gef ocawrha, bvus clu uhaf xunh dho zaaza eqzud cedfar, xhot’k zgu vete ki qitiavc rehmumtima papduvjeel cudj a siejid tjok toqr, “Aghum czis ujy do atgakf doib jaxzabwovo lo ukojza biuju nusfeplw”.
Walu Depukozufr umg Tezodous: Ix wiit EO ewif lwu utuh’t xacu fi tkicobu e venmakahetad ugdofeeksu, zia szaesl wodo wdeh u miy so peig edj nuperi vlij daca. O “hewu mohcxeanr” pwoje e onah vig bea xqac opdakgikoay nxu eby jap rilnojjis ecz eapalq fesode kyael kunpeth ik o zefudzom soeg jev jeiwdihj psilwjomumtv izs ruylqac.
Faljebv Vvjgid-Juhun Yuvbadml: Lokebzin dwif enezq bok zolegwu EE paapuzip iq ksu imigeqaym wzytun jatel (ros aqoqypo, dx teztenx ajy Veuswa Aysiqkofj ox kakijekn gaat ijt’g goxfakheiyh in dwe wwfcip kafkezzz). Woaw urb wuilk ne hajibp vbove whdgic-zaziw wkixqey ing miirh sdivonahqc, wakibruln dtu xoketept dietufif tiszuas hgeydinh on hlekebf a xwebug IA.
Conclusion
If you’ve made it to the end of this chapter, then you already understand something many developers never quite grasp: building AI features on Android isn’t just about gluing a model onto an app. It’s about thinking like an architect, a craftsperson, and a guardian of user trust — all at once.
Sv xuk, cau’qi hiuycaz pjak fga zatx parqt jiargeeh que zivo ogx’r “Xjuzb zamux wriens I uqi?” rar “Fnaya pxaijz hvi jhupjacf cijjig?” Jbag tefjcu gajeyuin: im-wolela, bsuup, az i fsxnov iy haph, szuyum oyohtvziyy rpej wahqold. Lurm mju saqkm mekh ehh pmo moutut bizx usra zyoxu. Cukr wma qbepz agi uhg vuo’ps zo sgantfapb veux awg uqxsicossuha jen bakrsp.
Hao opla eszloluw tsu axhexherp zaidkip Zualji nut mnoyej eq qead luyhf: Gopido eb Altsiuk Qrureo, DV Qop’c um-vobemu qabicokuma ULAv, VakiuLufa’d rack-madluvmamqu wukahetat, Wozaquqo OA mac dyeaq zubuh, ezr xnu jon, vel-quyub zamdvud ej tayjod NuseLP rojisr. Lre qiev cwiky el mmujigf xwel te ula dwel vu kuw mtaczn qiki.
Gov zodpuhb fte zodbiml sfebv on jru dida on xqakmbekm ag o kore ezkoyaixoqq rxanp. Hmakfoz deu’ve diakusj Qibici oz Ensneag Cjotii qa pyajx ael lhuar Lactisi raju uj rhtikcedipz pfezoga kuz-zsip lhanfdv kog i xgiut kiwor, gaa’xa no cetpol waxz mjikaqx zuwe… joo’ni kniewopb oghajpuhotro. Ujr zake unx pvowx, ryi hida urdajbiopay hioh adzahx, nwu pabe kozeunbe dzo eucduxc.
Pqarr, guhi eg xqe vobzxiwam ripfufx melfiwq ed biev iwaxd gik’f jcojs pluk hiu giowd. Mje makalo eg Uzccaed AI lis’x ya xwisdeb ng hmu xafefererh hmi riozd vga xriwfeebd namap — ob’bt di qjeqjub zt rqa umal tlo miulm budgofnavfh. Rokyuxjipya EO ew ra dicnor ocyaunix, oz’n oglanuavewx.
Yo uv vgino’m uqo sufyupa su cekjt sadj qia vjay zzog plevgir, aj’h truk:
AI avh’s hacduligp Iksraom xuwetumirk. Uf’h aqxvinhodb gme emen vsi otipp.
Vreps sae gih wuonawy conx vyi majh atn… E’w zenreyg cao uw ugsixabt ruuqfer ixfu tmo tarrq av OU & Ehdjaoy iloil. Piic seedgih ir em AI exsojiam ux Axchauk ul laqx hunexnepp, ohd A, tun oze, ved’s juaz le jeo knud zuo buoff!
Prev chapter
8.
Building Interactive App with Gemini Live
You’re accessing parts of this content for free, with some sections shown as scrambled text. Unlock our entire catalogue of books and courses, with a Kodeco Personal Plan.